SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2311.10263 · 2023

Stable Differentiable Causal Discovery

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2311.10263.json · badge: /api/ran/2311.10263.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
azizilab/sdcd canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
FunctionStatusWhere it lives
get_activation Ran
ran; our contract draft was wrong, not the code
azizilab/sdcd/sdcd/models/modules.py
code served (permissive licence) · get_code("60f197f4a335c446")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ReLU. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# The function takes a single parameter: activation (a string literal)
# It returns an nn.Module or callable based on the activation name
# We need to test with valid activation names: 'relu', 'sigmoid', 'tanh', 'linear'

ARGS = {
    'activation': 'relu'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ReLU — a ReLU output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ReLU output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
normalize Ran
ran; our contract draft was wrong, not the code
azizilab/sdcd/sdcd/models/modules.py
code served (permissive licence) · get_code("61555ad38527dd55")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 1, dtype family float, shape (4,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function normalize takes a vector v and normalizes it by dividing by its norm
# v should be a torch tensor
v = torch.randn(4)

ARGS = {
    "v": v
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (4,) — shape (4,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
scale_mechanism Ran
ran; our contract draft was wrong, not the code
azizilab/sdcd/sdcd/causal_model/mechanisms.py
code served (permissive licence) · get_code("082a62efb2cd4e05")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned TransformedDistribution. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.distributions as dist
import copy
from torch.distributions import AffineTransform, Distribution, TransformedDistribution

torch.manual_seed(0)

# Create a simple Distribution instance
simple_distribution = dist.Normal(loc=0.0, scale=1.0)

# Create a ConditionalDistribution subclass
class SimpleConditionalDistribution(dist.Distribution):
    """A simple conditional distribution for testing"""
    
    def __init__(self):
        super().__init__()
        self.parents = ["parent_var"]
        self.response_distribution_constructor = self._make_response_dist
    
    def _make_response_dist(self, **kwargs):
        """Construct a response distribution given parent values"""
        return dist.Normal(loc=0.0, scale=1.0)
    
    def compute_distribution_from_parents(self, parents_values):
        """Return distribution conditioned on parent values"""
        return self.response_distribution_constructor(**parents_values)
    
    def get_parents(self):
        """Return list of parent variable names"""
        return self.parents
    
    def sample(self, sample_shape, parents_values):
        """Sample from the conditional distribution"""
        return self.compute_distribution_from_parents(parents_values).sample(sample_shape)
    
    @property
    def mean(self):
        return torch.tensor(0.0)
    
    @property
    def variance(self):
        return torch.tensor(1.0)

conditional_distribution = SimpleConditionalDistribution()

# Test both paths: Distribution and ConditionalDistribution
ARGS = {
    "mechanism": simple_distribution,
    "scale": 0.5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type TransformedDistribution — a TransformedDistribution output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a TransformedDistribution output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Inferring causal relationships as directed acyclic graphs (DAGs) is an important but challenging problem. Differentiable Causal Discovery (DCD) is a promising approach to this problem, framing the search as a continuous optimization. But existing DCD methods are numerically unstable, with poor performance beyond tens of variables. In this paper, we propose Stable Differentiable Causal Discovery (SDCD), a new method that improves previous DCD methods in two ways: (1) It employs an alternative constraint for acyclicity; this constraint is more stable, both theoretically and empirically, and fast to compute. (2) It uses a training procedure tailored for sparse causal graphs, which are common in real-world scenarios. We first derive SDCD and prove its stability and correctness. We then evaluate it with both observational and interventional data and on both small-scale and large-scale settings. We find that SDCD outperforms existing methods in both convergence speed and accuracy and can scale to thousands of variables. We provide code at https://github.com/azizilab/sdcd.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2311.10263")
get_code_for_paper("2311.10263")
have("2311.10263")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2311.10263.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2311.10263.svg)](https://syntology.ai/paper/2311.10263)

Connect an agent — have() is free.